Principles of Protein Structure

Amino Acids - Basic Building Blocks of Proteins

Amino Acids Have

Molecular ChiralityMolecular Chirality

Proteins are Polymers

Cis Proline in the Active Site of HCV NS2

Glu 163

His 143 Leu 217(C-term)

Cys 184

3.2 Å3.0 Å

3.1 Å

4.1 Å

Cis Pro 164

3.1 Å

3.0 Å

Nature 2006 vol. 442 (7104) pp. 831-5

Cis trans Proline Isomerase• Cyclophilins are a family of proteins that

catalyze the isomerization of peptide backbones at a proline

• Critically important for protein folding • Cyclospronine is a potent

immunosuppressant and an inhibitor of cyclophilins

• Commonly used after organ transplant

Four Levels of Protein Structure

Four Levels of Protein Structure •  Primary, 1o!

–  Amino acid sequence; Covalent bonds"

•  Secondary, 2o!

–  Local conformation of main-chain atoms (Φ and Ψ angles); non-covalent interactions (h-bonds)"

•  Tertiary, 3o!

–  3-D arrangement of all the atoms in space (main-chain and side-chain); non-covalent interactions"

•  Quaternary, 4o!

–  3-D arrangement of subunit chains, non-covalent interactions

Four Levels of Protein StructureFour Levels of Protein Structure

•  Primary, 1o "

TPEEKSAVTALWGKV"•  Secondary, 2o!

•  Tertiary, 3o!

•  Quaternary, 4o!β2

β2 β1

α1 α2

Side Chain Conformation

Protein Backbone Conformation

α Helices

α Helix• If N-terminus is at bottom,

then all peptide N-H bonds point “down” and all peptide C=O bonds point “up”.

• N-H of residue n is H-bonded to C=O of residue n+4.

• a-Helix has:– 3.6 residues per turn– Rise/residue = 1.5 Å– Rise/turn = 5.4Å

Secondary Structure: Alpha HelicesRight handed Left handed

310 and π Helices

310 helix H-bonds n and n+3π helices H-bonds n and n+5

Polar Hydrophobic Amphipathic

α Helix• R-groups in α-helices:

– extend radially from the core, – shown in helical wheel diagram.– Can have varied distributions

Secondary Structure: Beta Sheet

Antiparallel beta sheet Parallel beta sheet

β Sheet

• Stabilized by H-bonds between N-H & C=O from adjacent stretches of strands

• Peptide chains are fully extended pleated shape because adjacent peptides groups can’t be coplanar.

ParallelNot optimum H-bonds; less stable

Anti-parallelOptimum H-bonds; more stable

β Sheet - 2 Orientations

Beta Turn – 2 Conformations

Only Difference

Tertiary Structure• Charge based interactions

– 62R:163E– 55E:170R

• Hydrophobic interactions– 189V– 201L– 213I– 215L– 266L

• Disulfide bond– 203C:259C

1HSAPeptide bound to Class I MHC

Quaternary Structure

β2 β1

α1α2

PDB File

Inter-atomic distances:Internal coordinates: bond lengths

(distance between bonded atoms)

b12 = ((a2x–a1x )2+(a2y–a1y )

2 +(a2z–a1z )2)1/2

a1

a2

b12

a1 = (a1x, a1y, a1z) a2 = (a2x, a2y, a2z)

Atomic coordinates

Bond length

Interatomic Distances

Bonded distances don’t change much:Bond lengths: Chain A backbone

PDB file 3e7l

1.459±0.003 Å1.527±0.003 Å1.329±0.001 Å

Bond Lengths Remain Constant

Bond or “valence” angles:Internal coordinates: valence angles (complement of angle formed by successive chemical bonds)

b12"b23 = (a2x–a1x)(a3x–a2x)+(a2y–a1y)(a3y–a2y)+(a2z–a1z)(a3z–a2z)

a3

cos("–!123) = cos" cos!123 + sin" sin!123 = – cos!123

b12"b23 = b12b23 cos("–!123)

a1

a2

b12

b23

!123

cos!123 = –b12"b23/(b12b23)

Atomic coordinates

a1 = (a1x, a1y, a1z) a2 = (a2x, a2y, a2z) a3 = (a3x, a3y, a3z)

b12 = (a2x–a1x, a2y–a1y, a2z–a1z) b23 = (a3x–a2x, a3y–a2y, a3z–a2z)

Bond vectors

Scalar product

Bond Angles

Distribution of backbone angles:

Valence angles: Chain A backbone

PDB file 3e7l

112.0±1.5°116.4±0.5°121.7±0.7°

Distribution of Backbone Angles

Distribution of B-factors

Motifs, Topologies and Folds: β Sheet The arrangement of secondary structure elements that give rise to a folded entity

Motifs, Topologies and Folds: β Sheet

Motifs, Topologies and Folds: βαβ

Motifs, Topologies and Folds: α-helical

Motifs, Topologies and Folds: α-helical

Middle Domain of eIF4G - scaffold protein for translation initiation factors.

N

C

Protein Domains

• A folded unit of protein

• normally formed around a hydrophobic core

• Tend to be globular

• 150-250 amino acids

Protein Domains

Many Eukaryotic proteins consist of multiple domains

• Limited proteolysis can be used to experimentally determine domain limits.

Limited Trypsin Digestion in the Absence and Presence of dsRNA

Domain Swapping

Location of His 143, Glu 163, Cys 184

35Å

His

Glu

CysCysGlu

His

HIV Protease

Retroviral aspartic protease - dimerization forms a single active site

Structural Comparison • Structural comparison requires a 3-Dimension

search• Minimize the search by using just secondary

structural elements - Potential problem?• DALI server

http://ekhidna.biocenter.helsinki.fi/dali_server/

• PDBe/foldhttp://www.ebi.ac.uk/msd-srv/ssm/cgi-bin/ssmserver

• VASThttp://www.ncbi.nlm.nih.gov/Structure/VAST/

SCOP:Structural Classification Of Proteins

SCOP database classifies domains not entire proteins

CATH:Class, Architecture, Topology, Homologous Superfamily

How to classify proteins with same general arrangement of secondary structure but are connected differently?